Convolutional Neural Network for Classification for Indian Jewellery

7 Pages Posted: 12 Jun 2019

See all articles by Kumar Vaibhav

Kumar Vaibhav

Amity University, Rajasthan

Jagdish Prasad

Amity University, Rajasthan

Baldev Singh

Vivekanand Global University

Date Written: March 13, 2019

Abstract

The classification of jewellery involves the identification of a jewellery among several jewellery images. The field has application is e-commerce. In the present paper, a deep convolutional neural network has been trained in a fully supervised manner on a dataset of Indian jewellery images. The network would able to classify a given query image of a jewellery to the multiple categories of the pre-trained categories of the network with the respective probabilities of each category. A classification accuracy of 88% has been achieved with the network.

Keywords: Convolutional Neural Network (CNN), Inception V3, Supervised Learning, Image Classification

Suggested Citation

Vaibhav, Kumar and Prasad, Jagdish and Singh, Baldev, Convolutional Neural Network for Classification for Indian Jewellery (March 13, 2019). Proceedings of International Conference on Sustainable Computing in Science, Technology and Management (SUSCOM), Amity University Rajasthan, Jaipur - India, February 26-28, 2019, Available at SSRN: https://ssrn.com/abstract=3351805 or http://dx.doi.org/10.2139/ssrn.3351805

Kumar Vaibhav

Amity University, Rajasthan ( email )

Jaipur
Kalawar
Rajasthan, Rajasthan 303002
India

Jagdish Prasad

Amity University, Rajasthan ( email )

Kant
Kalawar
Jaipur, Rajasthan 303002
India

Baldev Singh (Contact Author)

Vivekanand Global University ( email )

Jaipur
India

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
391
Abstract Views
1,530
Rank
190,168
PlumX Metrics